The present invention relates to computing devices. More particularly, the present invention provides a system including a plurality of computer modules that can independently operate to provide backup capability, dual processing, and the like. Merely by way of example, the present invention is applied to a modular computing environment for desk top computers, but it will be recognized that the invention has a much wider range of applicability. It can be applied to a server as well as other portable or modular computing applications.
Many desktop or personal computers, which are commonly termed PCs, have been around and used for over ten years. The PCs often come with state-of-art microprocessors such as the Intel Pentium™ microprocessor chips. They also include a hard or fixed disk drive such as memory in the giga-bit range. Additionally, the PCs often include a random access memory integrated circuit device such as a dynamic random access memory device, which is commonly termed DRAM. The DRAM devices now provide up to millions of memory cells (i.e., mega-bit) on a single slice of silicon. PCs also include a high resolution display such as cathode ray tubes or CRTs. In most cases, the CRTs are at least 15 inches or 17 inches or 20 inches in diameter. High resolution flat panel displays are also used with PCs.
Many external or peripheral devices can be used with the PCs. Among others, these peripheral devices include mass storage devices such as a Zip™ Drive product sold by Iomega Corporation of Utah. Other storage devices include external hard drives, tape drives, and others. Additional devices include communication devices such as a modem, which can be used to link the PC to a wide area network of computers such as the Internet. Furthermore, the PC can include output devices such as a printer and other output means. Moreover, the PC can include special audio output devices such as speakers the like.
PCs also have easy to use keyboards, mouse input devices, and the like. The keyboard is generally configured similar to a typewriter format. The keyboard also has the length and width for easily inputting information by way of keys to the computer. The mouse also has a sufficient size and shape to easily move a curser on the display from one location to another location.
Other types of computing devices include portable computing devices such as “laptop” computers and the like. Although somewhat successful, laptop computers have many limitations. These computing devices have poor display technology. In fact, these devices often have a smaller flat panel display that has poor viewing characteristics. Additionally, these devices also have poor input devices such as smaller keyboards and the like. Furthermore, these devices have limited common platforms to transfer information to and from these devices and other devices such as PCs.
Up to now, there has been little common ground between these platforms including the PCs and laptops in terms of upgrading, ease-of-use, cost, performance, and the like. Many differences between these platforms, probably somewhat intentional, has benefited computer manufacturers at the cost of consumers. A drawback to having two separate computers is that the user must often purchase both the desktop and laptop to have “total” computing power, where the desktop serves as a “regular” computer and the laptop serves as a “portable” computer. Purchasing both computers is often costly and runs “thousands” of dollars. The user also wastes a significant amount of time transferring software and data between the two types of computers. For example, the user must often couple the portable computer to a local area network (i.e., LAN), to a serial port with a modem and then manually transfer over files and data between the desktop and the portable computer. Alternatively, the user often must use floppy disks to “zip” up files and programs that exceed the storage capacity of conventional floppy disks, and transfer the floppy disk data manually.
Another drawback with the current model of separate portable and desktop computer is that the user has to spend money to buy components and peripherals the are duplicated in at least one of these computers. For example, both the desktop and portable computers typically include hard disk drives, floppy drives, CD-ROMs, computer memory, host processors, graphics accelerators, and the like. Because program software and supporting programs generally must be installed upon both hard drives in order for the user to operate programs on the road and in the office, hard disk space is often wasted.
One approach to reduce some of these drawbacks has been the use of a docking station with a portable computer. Here, the user has the portable computer for “on the road” use and a docking station that houses the portable computer for office use.
Similar to separate desktop and portable computers, there is no commonality between two desktop computers. To date, most personal computers are constructed with a single motherboard that provides connection for CPU and other components in the computer. Dual CPU systems have been available through Intel's slot 1 architecture. For example, two Pentium II cartridges can be plugged into two “slot 1” card slots on a motherboard to form a Dual-processor system. The two CPU's share a common host bus that connects to the rest of the system, e.g. main memory, hard disk drive, graphics subsystem, and others. Dual CPU systems have the advantage of increased CPU performance for the whole system. Adding a CPU cartridge requires no change in operating systems and application software. However, dual CPU systems may suffer limited performance improvement if memory or disk drive bandwidth becomes the limiting factor. Also, dual CPU systems have to time-share the processing unit in running multiple applications. CPU performance improvement efficiency also depends on software coding structure. Dual CPU systems provide no hardware redundancy to help fault tolerance. In running multiple applications, memory and disk drive data throughput will become the limiting factor in improving performance with multi-processor systems.
The present invention generally relates to computer interfaces. More specifically, the present invention relates to an interface channel that interfaces two computer interface buses that operate under protocols that are different from that used by the interface channel.
Interfaces coupling two independent computer buses are well known in the art. A block diagram of a computer system utilizing such a prior art interface is shown in FIG. 5. In FIG. 5, a primary peripheral component interconnect (PCI) bus 505 of a notebook PC 500 is coupled to a secondary PCI bus 555 in a docking system 550 (also referred to as docking station 550) through high pin count connectors 501 and 502, which are normally mating connectors. The high pin count connectors 501 and 502 contain a sufficiently large number of pins so as to carry PCI bus signals between the two PCI buses without any translation. The main purpose for interfacing the two independent PCI buses is to allow transactions to occur between a master on one PCI bus and a target on the other PCI bus. The interface between these two independent PCI buses additionally includes an optional PCI to PCI bridge 560, located in the docking station 550, to expand the add on capability in docking station 550. The bridge 560 creates a new bus number for devices behind the bridge 560 so that they are not on the same bus number as other devices in the system thus increasing the add on capability in the docking station 550.
An interface such as that shown in FIG. 5 provides an adequate interface between the primary and secondary PCI buses. However, the interface is limited in a number of ways. The interface transfers signals between the primary and secondary PCI buses using the protocols of a PCI bus. Consequently, the interface is subject to the limitations under which PCI buses operate. One such limitation is the fact that PCI buses are not cable friendly. The cable friendliness of the interface was not a major concern in the prior art. However, in the context of the computer system of the present invention, which is described in the present inventor's (William W. Y. Chu's) application for “Personal Computer Peripheral Console With Attached Computer Module” filed concurrently with the present application on Sep. 8, 1998 and incorporated herein by reference, a cable friendly interface is desired for interfacing an attached computer module (ACM) and a peripheral console of the present invention. Furthermore, as a result of operating by PCI protocols, the prior art interface includes a very large number of signal channels with a corresponding large number of conductive lines (and a similarly large number of pins in the connectors of the interface) that are commensurate in number with the number of signal lines in the PCI buses which it interfaces. One disadvantage of an interface having a relatively large number of conductive lines and pins is that it costs more than one that uses a fewer number of conductive lines and pins. Additionally, an interface having a large number of conductive lines is bulkier and more cumbersome to handle. Finally, a relatively large number of signal channels in the interface renders the option of using differential voltage signals less viable because a differential voltage signal method would require duplicating a large number of signal lines. It is desirable to use a low voltage differential signal (LVDS) channel in the computer system of the present invention because an LVDS channel is more cable friendly, faster, consumes less power, and generates less noise, including electromagnetic interferences (EMI), than a PCI channel. The term LVDS is herein used to generically refer to low voltage differential signals and is not intended to be limited to any particular type of LVDS technology.
Thus, what is needed are computer systems that can have multiple computer modules. Each computer module has dedicated memory and disk drive, and can operate independently.